MACHINE LEARNING MODELS TO PREDICT MARKET MOVEMENTS BASED ON HISTORICAL PRICE DATA AND ECONOMIC INDICATORS

Abstract

In the ongoing economic environment, predicting stock market prices is a significant point. Accordingly, researchers are presently more leaned to look for new chances to conjecture the stock market. We examine machine learning methods utilized in stock market estimating. Determining the bearing of the stock market's price advancement could yield critical benefits. To predict the future price of the stock, sellers utilize particular request, which is the investigation of price by looking at the past prices. A particular investigation instrument called a moving average assists the broker with distinguishing examples and pinpoint significant price focuses for stock exchanges. The moving normal is an incidental effect as well as a pattern marker. Reactive result is a financial indicator that is only provided following a significant price shift. The purpose of this work is to apply machine learning techniques to a specialized pointer. The suggested model will use relapse on moving averages to reduce the exchange signal's idle time and eventually overcome its drawback. By forecasting the exchange signal provided by the moving averages, the model is able to predict the pattern's inversion.